Natural History of Anal Ulcerations in Pediatric-Onset Crohn's Disease: Long-Term Follow-Up of a Population-Based Study.

INTRODUCTION: Anal ulcerations are frequently observed in Crohn's disease (CD). However, their natural history remains poorly known, especially in pediatric-onset CD. METHODS: All patients with a diagnosis of CD before the age of 17 years between 1988 and 2011 within the population-based registry EPIMAD were followed retrospectively until 2013. At diagnosis and during follow-up, the clinical and therapeutic features of perianal disease were recorded. An adjusted time-dependent Cox model was used to evaluate the risk of evolution of anal ulcerations toward suppurative lesions. RESULTS: Among the 1,005 included patients (females, 450 [44.8%]; median age at diagnosis 14.4 years [interquartile range 12.0-16.1]), 257 (25.6%) had an anal ulceration at diagnosis. Cumulative incidence of anal ulceration at 5 and 10 years from diagnosis was 38.4% (95% confidence interval [CI] 35.2-41.4) and 44.0% (95% CI 40.5-47.2), respectively. In multivariable analysis, the presence of extraintestinal manifestations (hazard ratio [HR] 1.46, 95% CI 1.19-1.80, P = 0.0003) and upper digestive location (HR 1.51, 95% CI 1.23-1.86, P < 0.0001) at diagnosis were associated with the occurrence of anal ulceration. Conversely, ileal location (L1) was associated with a lower risk of anal ulceration (L2 vs L1 HR 1.51, 95% CI 1.11-2.06, P = 0.0087; L3 vs L1 HR 1.42, 95% CI 1.08-1.85, P = 0.0116). The risk of fistulizing perianal CD (pCD) was doubled in patients with a history of anal ulceration (HR 2.00, 95% CI 1.45-2.74, P < 0.0001). Among the 352 patients with at least 1 episode of anal ulceration without history of fistulizing pCD, 82 (23.3%) developed fistulizing pCD after a median follow-up of 5.7 years (interquartile range 2.8-10.6). In these patients with anal ulceration, the diagnostic period (pre vs biologic era), exposure to immunosuppressants, and/or anti-tumor necrosis factor did not influence the risk of secondary anoperineal suppuration. DISCUSSION: Anal ulceration is frequent in pediatric-onset CD, with nearly half of patients presenting with at least 1 episode after 10 years of evolution. Fistulizing pCD is twice as frequent in patients with present or past anal ulceration.

New Therapeutic Strategies Are Associated With a Significant Decrease in Colectomy Rate in Pediatric Ulcerative Colitis.

INTRODUCTION: We evaluated the impact of immunosuppressants (IS) and antitumor necrosis factor (TNF) introduction on long-term outcomes of ulcerative colitis (UC) in a large population-based pediatric-onset cohort. METHODS: All patients included in the EPIMAD registry with a diagnosis of UC made before the age of 17 years between 1988 and 2011 were followed up retrospectively until 2013. Medication exposure and disease outcomes were compared between 3 diagnostic periods: 1988 to 1993 (period [P] 1; pre-IS era), 1994 to 2000 (P2; pre-anti-TNF era), and 2001 to 2011 (P3; anti-TNF era). RESULTS: A total of 337 patients (female, 57%) diagnosed with UC were followed up during a median duration of 7.2 years (interquartile range 3.8-13.0). The IS and anti-TNF exposure rates at 5 years increased over time from 7.8% (P1) to 63.8% (P3) and from 0% (P1) to 37.2% (P3), respectively. In parallel, the risk of colectomy at 5 years decreased significantly over time (P1, 17%; P2, 19%; and P3, 9%; P = 0.045, P -trend = 0.027) and between the pre-anti-TNF era (P1 + P2, 18%) and the anti-TNF era (P3, 9%) ( P = 0.013). The risk of disease extension at 5 years remained stable over time (P1, 36%, P2, 32%, and P3, 34%; P = 0.31, P -trend = 0.52) and between the pre-anti-TNF era (P1 + P2, 34%) and the anti-TNF era (P3, 34%) ( P = 0.92). The risk of flare-related hospitalization at 5 years significantly increased over time (P1, 16%; P2, 27%; P3, 42%; P = 0.0012, P -trend = 0.0006) and between the pre-anti-TNF era (P1 + P2, 23%) and the anti-TNF era (P3, 42%) ( P = 0.0004). DISCUSSION: In parallel with the increased use of IS and anti-TNF, an important decline in the risk of colectomy in pediatric-onset UC was observed at the population level.

New Therapeutic Strategies Have Changed the Natural History of Pediatric Crohn's Disease: A Two-Decade Population-Based Study.

BACKGROUND & AIMS: We evaluated the impact of immunosuppressants (IS) and anti-tumor necrosis factor (TNF) introduction on Crohn's disease (CD) long-term outcomes in a large population-based, pediatric-onset cohort. METHODS: All patients included in the EPIMAD registry with a diagnosis of CD occurring when they were younger than age 17 years and between 1988 and 2011 were followed up retrospectively until 2013. Three diagnostic periods were defined: 1988 to 1993 (period [P]1; pre-IS era), 1994 to 2000 (P2; pre-anti-TNF era), and 2001 to 2011 (P3; anti-TNF era). Medication exposure and disease outcomes were compared between the 3 diagnostic periods. RESULTS: A total of 1007 patients diagnosed with CD were followed up for a median duration of 8.8 years (interquartile range, 4.6-14.2 y). The IS and anti-TNF exposure rate at 5 years increased over time from 33.9% (in P1) to 76.5% (in P3) and from 0% (in P1) to 50.5% (in P3), respectively. In parallel, the risk for intestinal resection at 5 years decreased significantly over time (P1, 35%; P2, 31%; and P3, 22%; P = .0003, Ptrend < .0001), and between the pre-anti-TNF era (P1 + P2, 32%) and the anti-TNF era (P3, 22%) (P = .0007). The risk for progression from inflammatory to stricturing behavior decreased significantly over time (P1, 27%; P2, 28%; and P3, 20%; P = .11, Ptrend = .041) and between the pre-anti-TNF era (P1 + P2, 28%) and the anti-TNF era (P3, 20%) (P = .040). The risk for a CD flare-related hospitalization at 5 years remained stable over time (P1, 31%; P2, 31%; and P3, 29%; P = .76, Ptrend = .53). CONCLUSIONS: In parallel with the increased use of IS and anti-TNF, positive changes in the natural history of pediatric-onset CD were observed at the population level. A decreased risk of both intestinal resections and stricturing complications were observed during the anti-TNF era.

Long-term effectiveness and safety of anti-TNF in pediatric-onset inflammatory bowel diseases: A population-based study.

BACKGROUND: Anti-TNF agents are the first biologic treatment option in inflammatory bowel disease (IBD). The long-term effectiveness of this strategy at the population level is poorly known, particularly in pediatric-onset IBD. METHODS: All patients diagnosed with Crohn's disease (CD) or ulcerative colitis (UC) before the age of 17 between 1988 and 2011 in the EPIMAD population-based registry were followed retrospectively until 2013. Among patients treated with anti-TNF, the cumulative probabilities of anti-TNF failure defined by primary failure, loss of response (LOR) or intolerance were evaluated. Factors associated with anti-TNF failure were investigated by a Cox model. RESULTS: Among a total of 1,007 patients with CD and 337 patients with UC, respectively 481 (48%) and 81 (24%) were treated with anti-TNF. Median age at anti-TNF initiation was 17.4 years (IQR, 15.1-20.9). Median duration of anti-TNF therapy was 20.4 months (IQR, 6.0-59.9). In CD, the probability of failure of 1st line anti-TNF at 1, 3 and 5 years was respectively 30.7%, 51.3% and 61.9% for infliximab and 25.9%, 49.3% and 57.7% for adalimumab (p = 0.740). In UC, the probability of failure of 1st line anti-TNF therapy was respectively 38.4%, 52.3% and 72.7% for infliximab and 12.5% for these 3 timepoints for adalimumab (p = 0.091). The risk of failure was maximal in the first year of treatment and LOR was the main reason for discontinuation. Female gender was associated with LOR (HR, 1.48; 95%CI 1.02-2.14) and with anti-TNF withdrawal for intolerance in CD (HR, 2.31; 95%CI 1.30-4.11) and disease duration (≥ 2 y vs. < 2 y) was associated with LOR in UC (HR, 0.37; 95%CI 0.15-0.94) in multivariate analysis. Sixty-three (13.5%) patients observed adverse events leading to termination of treatment (p = 0.57). No death, cancer or tuberculosis was observed while the patients were under anti-TNF treatment. CONCLUSION: In a population-based study of pediatric-onset IBD, about 60% in CD and 70% in UC experienced anti-TNF failure within 5 years. Loss of response account for around two-thirds of failure, both for CD and UC.

Increased Risk of Cancer and Mortality in a Large French Population-Based Paediatric-Onset Inflammatory Bowel Disease Retrospective Cohort.

BACKGROUND AND AIMS: Paediatric-onset IBD [pIBD] is associated with an increased risk of cancer and mortality in adulthood. The aims of this study were to measure the incidence of cancer and mortality in patients with pIBD and identify factors associated with mortality and cancer. METHODS: All patients diagnosed with Crohn's disease [CD] or ulcerative colitis [UC] before the age of 17 years between 1988 and 2011 in the EPIMAD registry were retrospectively followed until 2013 for cancer and 2015 for mortality. Standardized incidence [SIR] and mortality ratios [SMR] were estimated compared to the general population. Cox regression was used to compare the effect of exposures on cancer and mortality among IBD patients. RESULTS: We included 1344 patients [52% males, 75% CD], totalling 12 957 patient-years for cancer incidence and 18 817 patient-years for mortality. There were 14 cases of cancer [median age 27.8 years] and 15 deaths [median age 28.8 years]. The incidence of cancer and of mortality were increased compared to the general population: all-cancer SIR = 2.7 (95% confidence interval [CI]: 1.5-4.8), SMR = 1.7 [95% CI: 1.0-2.8]. Colorectal cancer had the highest SIR and SMR: SIR = 41.2 [95% CI: 17.2-99.0], SMR = 70.4 [95% CI 22.7-218.2]. Cancer was associated with (hazard ratio [HR], 95% CI): active smoking at diagnosis [5.5, 1.8-16.5], p = 0.002; any exposure to anti-tumour necrosis factor [6.1, 1.7-22.3], p = 0.0065; and exposure to combination therapy [7.4, 1.8-29.7], p = 0.0047. Mortality was associated with extraintestinal manifestations (HR 4.9 [95% CI: 1.7-13.8], p = 0.003). CONCLUSIONS: In this large population-based cohort, patients with pIBD had an increased risk of both cancer [2.7-fold] and mortality [1.7-fold], particularly for colorectal cancer.